The concept of genetic colinearity between the DNA and protein sequences has been found to be violated in many genes during the last ten years. The most obvious example is RNA splicing. Many other examples exist involving unusual mechanisms of transcription and translation. Each of these unusual modes of expression provide molecular biology with new tools to study fundamental genetic processes. We are studying an unusual very efficient ribosomal frameshift which occurs in the Ty elements of the yeast Saccharomyces cerevisiae. The Ty elements of yeast are a dispersed family of 30 related retroviral-like transposons. The elements include two genes, TYA and TYB, which are analogous to gag and pol of retroviruses. TYB is expressed as a fusion to the TYA gene. This fusion could occur by splicing or RNa editing. However since the RNA sequence of the overlap between the two genes is colinear with the DNA only translational models explain TYB expression. We refer to them generically as frameshifting, realizing that mechanism has not been defined. A 14 nucleotide sequence is both necessary and sufficient to promote 20% frameshifting; and that it is the site of the frameshift. This sequence includes no obvious secondary structure, but does include an unusual codon, AGG. This codon is recognized by a rare tRNA encoded by a single genetic locus. It may be that a translational pause induced by the low abundance of the tRNA is required for the frameshift to occur. Many questions remain unanswered about Ty site-specific translational frameshifting. What nucleotides are essential to the process? Would overproduction of the cognate tRNA eliminate frameshifting dependent upon the AGG codon? Could other codons recognized by rare tRNAs substitute? What external factors (ribosomal proteins? translation factors?) regulate the event? Is translational termination required immediately distal to the frameshift site? Frameshifting promoted by the 14 nucleotide minimal site is suppressed near the TYA initiator AUG. This suppression may be caused by the context surrounding the site, though this seems unlikely. Alternatively the early steps in elongation could be different in some way which precludes frameshifting. Frameshifting may operate in a frame-independent way, though most efficient when occurring between the TYA and TYB frames. This result is difficult to reconcile with a model which requires a specific codon-anticodon interaction at the frameshift site. Biochemical analysis can best demonstrate the site of the Ty frameshift. We are purifying the product of a chimeric gene in which a portion of the S. aureus protein A gene is expressed dependent upon the 14 nucleotide minimal site. The sequence of a cyanogen bromide fragment encompassing the frameshift site and protein A fragment will be determined on a gas-phase sequenator. We are also using in vitro translation to demonstrate that an RNA colinear with the DNA can allow TYB expression.